AI System Performance Engineer

Job Title

AI System Performance Engineer

Role Summary

Design and validate analytical models and simulator integrations to project end-to-end performance of AI training and inference systems on Intel architectures. Identify architectural bottlenecks across GPU cores, memory hierarchy, interconnect, and communication strategies, and recommend optimal system configurations.

Work is collaborative and hands-on, covering multi-GPU deployment, parallelism, collective communication, and simulator/model adaptation for Intel-specific architecture. This role is eligible for a hybrid work model in Bangalore, India.

Experience Level

Entry-level - College graduate or 1+ years of relevant experience in AI workload and GPU architecture modeling/simulation or a closely related field.

Responsibilities

Key responsibilities include modelling, simulation, and performance analysis for AI workloads and GPU architectures.

• Enable new AI workloads in analytical simulators and extend simulator capabilities for multi-GPU deployment and communication strategies.
• Build analytical models of GPU architectures (core, memory hierarchy) and training/inference deployment scenarios to produce performance estimates and identify bottlenecks.
• Integrate and adapt open-source simulators or models for Intel-specific architectures.
• Perform parametric what‑if analyses across core counts, memory/cache capacity and bandwidth, fabric bandwidth, GPU interconnect, and deployment options.
• Deliver optimization recommendations and advice on optimal architecture and system configurations.

Requirements

Required technical skills and experience for performing the role.

Must-have:

• 1+ years of experience in AI workload and GPU architecture modelling/simulation or similar engineering experience.
• Strong proficiency in Python for building modular analytical models and simulation frameworks.
• Deep understanding of computer architecture: GPU/CPU components (ALU/SM), cache, memory types, and hierarchy.
• Knowledge of LLM AI workloads and kernel internals (Transformer, MoE, Attention; examples cited: LLaMA, DeepSeek, Qwen) and related runtime behaviors.
• Ability to analyze scaling/communication patterns for multi‑GPU training and inference.

Nice-to-have:

• Familiarity with distributed training/inference strategies (data, pipeline, tensor, expert parallelism) and their communication patterns.
• Understanding of LLM inference/training mechanics (prefill vs decode, KV cache, batching policies, quantization).
• Exposure to TRT-LLM, vLLM, or SGLang.
• Familiarity with Git/GitHub, CI/CD, and automated workflows.
• Background in power fundamentals and power/performance co-optimization.

Education Requirements

Required: BTech or Master’s degree in Electrical Engineering, Computer Engineering, Computer Science, or a related technical field; college graduate status is expected. (The posting indicates college graduate or 1+ years of experience.)

About the Company

Company: Intel Corporation

Headquarters: Santa Clara, California, USA

Intel Corporation is a leading multinational technology company known for its innovative semiconductor solutions, including microprocessors, artificial intelligence accelerators, and memory products. Headquartered in the United States, Intel focuses on cutting-edge technology and a collaborative working environment, driving advancements in semiconductor manufacturing to meet global demands. The company emphasizes professional development and aims to shape the future of technology through groundbreaking designs.